Theoretically Principled Trade-off between Robustness and Accuracy

By Hongyang Zhang, Yaodong Yu, Jiantao Jiao, Eric P. Xing, Laurent El Ghaoui, Michael I. Jordan. 2019

3,092 citations recorded by the source corpus.

We identify a trade-off between robustness and accuracy that serves as a guiding principle in the design of defenses against adversarial examples. Although this problem has been widely studied empirically, much remains unknown concerning the theory underlying this trade-off. In this work, we decompose the prediction error for adversarial examples (robust error) as the sum of the natural (classification) error and boundary error, and provide a differentiable upper bound using the theory of classification-calibrated loss, which is shown to be the tightest possible upper bound uniform over all probability distributions and measurable predictors. Inspired by our theoretical analysis, we also design a new defense method, TRADES, to trade adversarial robustness off against accuracy. Our proposed algorithm performs well experimentally in real-world datasets. The methodology is the foundation of our entry to the NeurIPS 2018 Adversarial Vision Challenge in which we won the 1st place out of ~2,000 submissions, surpassing the runner-up approach by $11.41\%$ in terms of mean $\ell_2$ perturbation distance.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering, Hardware and Architecture, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

  1. Percy Liang
  2. Mohit Bansal
  3. Jianfeng Gao
  4. Yoshua Bengio
  5. Quoc V. Le
  6. Sergey Levine
  7. Ruslan Salakhutdinov
  8. Dacheng Tao

In Electrical and Electronic Engineering

  1. Matthias Hein
  2. Tianyu Pang
  3. Percy Liang
  4. Maksym Andriushchenko
  5. Sven Gowal
  6. Francesco Croce
  7. Timothy A. Mann
  8. Jun Zhu

In Hardware and Architecture

  1. Matthias Hein
  2. Huan Zhang
  3. Cho-Jui Hsieh
  4. Tom Goldstein
  5. Francesco Croce
  6. Bernt Schiele
  7. Nicholas Carlini
  8. David Stutz

In Safety, Robustness & Interpretability

  1. Matthias Hein
  2. Nicholas Carlini
  3. Tianyu Pang
  4. Cho-Jui Hsieh
  5. Huan Zhang
  6. Francesco Croce
  7. Jun Zhu
  8. Percy Liang

Citations in the corpus